Superior productivity of American industrial workers

[Tom from Cornwall]

That's not an export, AFAICS. No more than is the tank used in Russia.

Sorry, didn't make myself clear enough. I thought that some of the statistics of the German war economy included figures for exports that someone had quoted in one of these mega-threads and I just wondered if there was any indication of what they were, where they went and when. Hence my examples.

I'll see if I can rustle up something myself though.

Regards

Tom

[KDF33]

Elsewhere, Richard gives us the total number of aircraft industry workers in each country:

Okay, so then in 3Q43, the "final process of assembly" labor in the aviation industry (airframe, engines, and props) in Germany was 935,000 and the US was 910,851 (airframe), 296,949 (engines), and 157,593 (props), for a total of 1,365,393. Now do the productivity calculation.

He doesn't. There are two problems with the comparison: first, the German figure of 935,000 workers, sourced from here, comprises a wider category of workers than those producing 'airframes, engines and propellers'. Namely, it includes workers producing 'Aircraft equipment' - presumably radios, weapons, etc. Second, the figure of 910,851 U.S. workers engaged in airframe production differs from the USSBS figure of 1,084,000 for 5/31/43. Presumably, the USSBS knew its U.S. data. The simplest explanation? Richard's figure excludes subcontractors:

If we use Richard's figures, then I get:

A false impression. I've actually reconciled the two contradictory sets of USSBS figures. The correct figures are found here, in pounds. In this other report, the figures are given as metric ones, a confounding mistake.

The way the USSBS derived German airframe weights must be qualified: their figures aren't official German data, but rather estimates made by the authors, using unchanging airframe weights for each German aircraft type throughout the war years, not taking account of variance in weight between models, which tended to get heavier as the war progressed. It is left unsaid whether U.S. figures are calculated on the same basis, but given that they come from an official U.S. digest, I would tend to assume that they are more precise.

In any event, German figures include some small production outsourced to occupied countries, so at the end of the day, one slight discrepancy in one direction might well be corrected by another in the other direction.

On that basis, I have created a table to compare German and U.S. (without U.S. orders in Canada) production of airframes and parts:



Data is in pounds. It includes spare parts for both countries. The column 'Ratio' is German production expressed as a percent of that of the U.S. The column 'Bomb t.' is the tonnage (short tons) dropped on aircraft production targets in Europe by Allied strategic air forces, and is sourced from here.

Beneath the blue line (row 18) is the USSBS data on manpower (including subcontractors) employed on airframe production, for both countries on the 31st of May of each year. Cases E19 to E21 are German figures expressed as a percent of the U.S. Cases G19 to G21 are what most interests us, and show productivity.

They are calculated thus, for each year:

German production for May + June / workforce on May 31st, divided by U.S. production for May + June / workforce on May 31st.

We see that German productivity was 65% of the U.S. in spring 1942 and 70.5% in spring 1943. This, however, would already constitute degraded figures due to (1) the replacement of German with foreign workers (most prominent in 1942) and, starting in the second quarter of 1943, (2) strategic bombing on aircraft production targets. By spring 1944, the effects of strategic bombing appear to have led to a collapse of German productivity, which by then was no more than 44.4% of the U.S.

If we correct for the negative effects of slave labor and early strategic bombing on German productivity, I would estimate 'real' German productivity in the aircraft industry at ~80% of the U.S.

[TheMarcksPlan]

Nice work.

Data is in pounds.

Raw pounds? Or did you adjust, as USSBS did in Figure VI, to fighter-equivalents based on frame weight?

USSBS Figure VI derives fighter-equivalents based on the cube root of weight, which is too big an adjustment by the US per-pound stats I created here.

It includes spare parts for both countries.

Did you create your own data, as with empty weight, or did you use USSBS figures?

I would estimate 'real' German productivity in the aircraft industry at ~80% of the U.S.

You're making a good case. The remaining sources of uncertainty are (1) proportion of imported German production whose labor isn't represented and (2) true size of German (and American) labor force. I think you're right that (1) is likely to be smallish. On (2) who's to say which stats are more or less skewed, maybe we assume unaccounted labor force is similar in proportion.

If your data are for raw pounds, some of the '43-'44 difference in differences chalks up to German switch to fighters.

[KDF33]

Nice work.

Thank you!

Raw pounds? Or did you adjust, as USSBS did in Ex. 5, to fighter-equivalents based on frame weight?

Unadjusted raw pounds, yes.

USSBS Ex. 5 derives fighter-equivalents based on the cube root of weight, which is too big an adjustment by the US per-pound stats I created here.

The USSBS fighter-equivalent figures don't really make sense. For 1944, the adjustments reduce the American figure by 21%, but also reduce the German figure by 31%!

Did you create your own data, as with empty weight, or did you use USSBS figures?

I used the USSBS for Germany, found here. Now that I've reconciled the different sets from the various reports I'm confident that it can be used. For the U.S. I used the figures found in Official Munitions Production of the United States - which is what the USSBS uses, when it cites U.S. figures.

You're making a good case. The remaining sources of uncertainty are (1) proportion of imported German production whose labor isn't represented and (2) true size of German (and American) labor force. I think you're right that (1) is likely to be smallish. On (2) who's to say which stats are more or less skewed, maybe we assume unaccounted labor force is similar in proportion.

At least in terms of airframe assembly, (1) shouldn't make much of a difference. In terms of the size of the labor force, it is clear that the USSBS data is only for airframes, and doesn't include engines, propellers or any related aircraft equipment. It is clear on including subcontractors for both countries, so at least for that specific subset of aircraft production workforce it seems to be an apples-to-apples comparison.

If your data are for raw pounds, some of the '43-'44 difference in differences chalks up to German switch to fighters.

Do you know of any good source that details to what extent fighters are more expensive on a per-weight basis than bombers? Also, why is the USSBS data seemingly suggesting that it is the other way around?

[TheMarcksPlan]

Do you know of any good source that details to what extent fighters are more expensive on a per-weight basis than bombers? Also, why is the USSBS data seemingly suggesting that it is the other way around?

I don't know such a source. From past professional experience with the airliner industry, however, I know that certain expensive components - avionics, navigation, climate monitoring - scale very little with frame size. It's one reason that small airliners like CRJ-200 and Do-328 are no longer produced. Intuitively, same would apply to WW2 planes - can't see why not. P-51's pitot tubes need to be just as accurate as B-17's. The electric/hydraulic leads to control surfaces are very expensive equipment and, though they scale somewhat with weight, not perfectly (linear plumbing length vs. quadratic/cubic escalation of control surface area and weight).

Additionally, the proportion of engine cost should be significantly higher for 1E fighters. Again that's just intuition but seems sort of obvious. To the extent USSBS labor figures exclude engine/propeller labor, that skews the analysis in favor of Germany (at least for the later war, earlier the LW is bomber-heavy).

The USSBS fighter-equivalent figures don't really make sense. For 1944, the adjustments reduce the American figure by 21%, but also reduce the German figure by 31%!

Didn't notice that. Has to be a spreadsheet error. One problem with USSBS is that, even with a massive workforce, they didn't have time to correct all errors and refine their analysis. Here we are 80 years later (and many smarter people as well), still trying to get the picture right.

[TheMarcksPlan]

It is clear on including subcontractors for both countries, so at least for that specific subset of aircraft production workforce it seems to be an apples-to-apples comparison.

There's subcontractors hired usually to build specific bespoke sub-assemblies, such as Adam Opel building the Ju-88's control surfaces.

But then there's firms who build masses of generic parts - standardized high-tolerance screws (planes have thousands of them), generic small forged pieces that make up sub assemblies. These are probably purchased instead of sub-contracted.

Going back to Tooze's characterization of Wagenfuhr's labor table in Statistics and the German State:

WagenfuÈhr was able to include sub-contracted specialized
components, such as the labour and steel that went into casting the hull
of a tank. However, his table did not record any of the resources that
went into the mass of generic components from which the tank was
assembled.

I'm guessing the labor behind generic components is as absent from USSBS reports as it was from Wagenfuhr's Gesamtplan - USSBS relied on him a lot, after all.

I'm somewhat suspicious that German planemaking involved more generic components than American because the latter were, generally, more sophisticated. The extent of out-sourcing in the German aviation industry is also remarkable, with final-assembly labor being only 4.4% of Ju-88 production cost in one case:



American planemakers may have been similarly outsourced but, given the massive scale and vertical integration of firms like Ford and Boeing, I doubt it. This differential may have extended all the way down: e.g. Ford might have made its own high-tolerance screws/rivets while Junkers and their sub-contractors may have bought them. That's just a hunch.

[Richard Anderson]

He doesn't. There are two problems with the comparison: first, the German figure of 935,000 workers, sourced from here, comprises a wider category of workers than those producing 'airframes, engines and propellers'. Namely, it includes workers producing 'Aircraft equipment' - presumably radios, weapons, etc. Second, the figure of 910,851 U.S. workers engaged in airframe production differs from the USSBS figure of 1,084,000 for 5/31/43. Presumably, the USSBS knew its U.S. data. The simplest explanation? Richard's figure excludes subcontractors:

Presumably Wartime Development of the Aircraft Industry: Bulletin of the United States Bureau of Labor Statistics, No. 800 knew its data too.

Total number of prime contractor employees in the aircraft industry as of 5/31/43 was 1,211.6 million, subs and "parts suppliers" were 625,000. However, the numbers of prime contractor employees in the airframe plants was 856,244, the engine plants was 263,684, and propeller plants was 46,627, for a total of 1,166,555.

Subcontractors and parts suppliers included "employment in many plants classified by the Bureau's Employment Statistics Division in other industries, such as electrical equipment and automobiles; all establishments having sub­contracts are included, even when aircraft and parts do not constitute their primary activity."

[TheMarcksPlan]

American planemakers may have been similarly outsourced but, given the massive scale and vertical integration of firms like Ford and Boeing, I doubt it. This differential may have extended all the way down: e.g. Ford might have made its own high-tolerance screws/rivets while Junkers and their sub-contractors may have bought them. That's just a hunch.

Just to put a pin in this point:

Were the foregoing true, Ford's labor force (therefore US aviation industry) would include screw/rivet/bolt-making labor, while the German would not.

[historygeek2021]

He doesn't. There are two problems with the comparison: first, the German figure of 935,000 workers, sourced from here, comprises a wider category of workers than those producing 'airframes, engines and propellers'. Namely, it includes workers producing 'Aircraft equipment' - presumably radios, weapons, etc. Second, the figure of 910,851 U.S. workers engaged in airframe production differs from the USSBS figure of 1,084,000 for 5/31/43. Presumably, the USSBS knew its U.S. data. The simplest explanation? Richard's figure excludes subcontractors:

A false impression. I've actually reconciled the two contradictory sets of USSBS figures. The correct figures are found here, in pounds. In this other report, the figures are given as metric ones, a confounding mistake.

The way the USSBS derived German airframe weights must be qualified: their figures aren't official German data, but rather estimates made by the authors, using unchanging airframe weights for each German aircraft type throughout the war years, not taking account of variance in weight between models, which tended to get heavier as the war progressed. It is left unsaid whether U.S. figures are calculated on the same basis, but given that they come from an official U.S. digest, I would tend to assume that they are more precise.

How did you get your data for aircraft weight? Did you go through all the tables in the USSBS Aircraft Division Industry Report, look up the weight of each aircraft, and add them up? If so, that is very impressive. Thank you for your work.

Presumably Wartime Development of the Aircraft Industry: Bulletin of the United States Bureau of Labor Statistics, No. 800 knew its data too.

Total number of prime contractor employees in the aircraft industry as of 5/31/43 was 1,211.6 million, subs and "parts suppliers" were 625,000. However, the numbers of prime contractor employees in the airframe plants was 856,244, the engine plants was 263,684, and propeller plants was 46,627, for a total of 1,166,555.

Subcontractors and parts suppliers included "employment in many plants classified by the Bureau's Employment Statistics Division in other industries, such as electrical equipment and automobiles; all establishments having sub­contracts are included, even when aircraft and parts do not constitute their primary activity."

If we compare Richard's labor figures with KDF's, I get the following chart:

WW2 Aircraft Productivity USA v GER.png

My numbers come out the same as KDF when using his employment data.

Richard provides data for only a single point in time (mid-1943), so I interpolated the data in his side of the chart for 1942 and 1944 based on trends in employment in both countries (Germany increasing slightly each year, the US increasing drastically from 1942 to 1943 and then declining slightly in 1944).

It's also interesting to note that Richard gives a figure of 625,000 U.S. "equipment" workers, and KDF's data suggest that Germany had a roughly equal amount of equipment workers. This doesn't make a lot of sense, given that the USA employed more than twice as many workers in direct aircraft components and assembly. You would think that Germany would have far fewer equipment workers than the USA too.

In any event, both KDF's and Richard's data show that U.S. productivity far exceeded that of Germany. When we take equipment workers into account (in Richard's data), the U.S. productivity advantage is significantly higher. Moreover, using either data set, the productivity gap widened significantly in 1944.

The USSBS report (p. 146) states that the decline in German output due to air raids in 1944 can be estimated at 18%. I've added a fourth row ("No Air Raids") under each set of data to show what German output and productivity would have been without air raids, according to the USSBS.

The other big factor is that U.S. productivity soared in 1944 relative to 1943. German productivity rose each year, but not nearly as fast as U.S. productivity.

[Boby]

All this productivity debate is interesting but flawed imho, it gives a misleading impression that US and German aircraft industries were equal in specialized machine-tools, floor space and raw materials. Also USSBS weight output is more than questionable.

If you can exchange 373,000 US labor for that of Germany in 1943 (per KDF33 figure), would they be more productive in german factories?

[historygeek2021]

If you can exchange 373,000 US labor for that of Germany in 1943 (per KDF33 figure), would they be more productive in german factories?

We're not asking whether American workers were superior human beings or workers compared to German workers. We're asking by how much the output per worker in the United States exceeded the output per worker in Germany during the war. A number of factors affect that, not only the skill, motivation, strength and health of the workers, but also the equipment they're using, how much raw material they have, the size, layout and proximity of their factories, and the availability of fuel, vehicles, trains and planes to transport workers, components and products where they need to be, etc.

[Boby]

If you can exchange 373,000 US labor for that of Germany in 1943 (per KDF33 figure), would they be more productive in german factories?

We're not asking whether American workers were superior human beings or workers compared to German workers. We're asking by how much the output per worker in the United States exceeded the output per worker in Germany during the war. A number of factors affect that, not only the skill, motivation, strength and health of the workers, but also the equipment they're using, how much raw material they have, the size, layout and proximity of their factories, and the availability of fuel, vehicles, trains and planes to transport workers, components and products where they need to be, etc.

Well, here you are introducing only labor and output.

Another factor (among many) was the number of/and new models, etc.

Anyway, what conclusions can we drawn if german factories, overall, were 20 or 30% less productive at their peak than US factories?

[historygeek2021]

If you can exchange 373,000 US labor for that of Germany in 1943 (per KDF33 figure), would they be more productive in german factories?

We're not asking whether American workers were superior human beings or workers compared to German workers. We're asking by how much the output per worker in the United States exceeded the output per worker in Germany during the war. A number of factors affect that, not only the skill, motivation, strength and health of the workers, but also the equipment they're using, how much raw material they have, the size, layout and proximity of their factories, and the availability of fuel, vehicles, trains and planes to transport workers, components and products where they need to be, etc.

Well, here you are introducing only labor and output.

Another factor (among many) was the number of/and new models, etc.

Anyway, what conclusions can we drawn if german factories, overall, were 20 or 30% less productive at their peak than US factories?

The thread topic is labor productivity. Labor productivity = output / labor

The obvious conclusion of this thread is that the U.S. industrial workforce was far more productive than the German industrial workforce, and that therefore, even if Germany had been able to increase the size of its industrial labor force, it never could have come close to matching U.S. industrial output, not to mention the output of the rest of the Allied countries.

[historygeek2021]

There's also the question of hours per laborer. If Germany extended its working day to 10 hours and required its workers to work 6 or 7 days a week, while American workers only worked 8 hours 5 days a week, then America's productivity advantage was even higher than shown in this thread.

[Boby]

If you can exchange 373,000 US labor for that of Germany in 1943 (per KDF33 figure), would they be more productive in german factories?

We're not asking whether American workers were superior human beings or workers compared to German workers. We're asking by how much the output per worker in the United States exceeded the output per worker in Germany during the war. A number of factors affect that, not only the skill, motivation, strength and health of the workers, but also the equipment they're using, how much raw material they have, the size, layout and proximity of their factories, and the availability of fuel, vehicles, trains and planes to transport workers, components and products where they need to be, etc.

Well, here you are introducing only labor and output.

Another factor (among many) was the number of/and new models, etc.

Anyway, what conclusions can we drawn if german factories, overall, were 20 or 30% less productive at their peak than US factories?

The thread topic is labor productivity. Labor productivity = output / labor

The obvious conclusion of this thread is that the U.S. industrial workforce was far more productive than the German industrial workforce, and that therefore, even if Germany had been able to increase the size of its industrial labor force, it never could have come close to matching U.S. industrial output, not to mention the output of the rest of the Allied countries.

I agree. My previous comment was only about the reasons of this superior productivity, nothing more.

Boby,